cs.SDNov 9, 2026

MUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice Protection

Authors: Donghyun Kim, Jin Hong, Seungmin Kim, Dain Kim, Junseok Kwon, Daeseon Choi

Organizations: Soongsil University · Chung-Ang University · AIM Intelligence

Abstract

Talking-head generation models, which synthesize realistic facial animations from audio, are increasingly vulnerable to misuse in multimodal deepfake scenarios. Protecting such systems remains challenging, as talking-head models fundamentally rely on precise audio-visual alignment, while effective audio protection methods remain largely underexplored. In this work, we propose MUTE, an audio protection framework that uncouples the underlying audio-visual alignment through a multi-level strategy. Specifically, MUTE combines (1) Alignment-level Disruption, which directly perturbs cross-modal attention to disrupt structural alignment, and (2) Representation-level Degradation, which perturbs temporal audio embeddings to degrade their representations. To improve robustness, perturbations are constrained in the STFT domain to high-energy regions, making them resistant to post-processing, denoising, and adaptive attacks. An optional speaker-level objective further mitigates potential bypass via TTS-based resynthesis. Extensive experiments demonstrate that MUTE consistently degrades lip synchronization across both white-box and black-box talking-head models while preserving perceptual audio quality. The proposed method remains effective under real-world transformations and can be combined with image-based approaches to provide stronger multimodal protection. The source code and project page are available at https://github.com/jindl465/MUTE.

Figures & tables

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Adaptive Compression and Targeted Perturbation: A Unified Framework for Generalized Audio Deepfake Detection

    Oct 12, 2026Jianqiao Cui, Haoxian Zhu, Hanbiao Meng +1Audio Deepfake DetectionAudio Deepfake Detection Robustness

  2. Audio-Driven Adversarial Defense for 3D Talking Face Generation with totally Visual Fidelity Preservation

    Aug 31, 2026Rui-Qing Sun, Chen-Hao Cui, Hui-Yang Zhao +3Audio-Driven Facial AnimationAdversarial Examples